Background: Pancreatic adenocarcinoma (PAAD) is an aggressive and invasive tumor with poor prognosis. Identifying prognostic biomarkers of PAAD will provide crucial information for developing treatment plans. Methods: In this analysis, a gene-expression dataset, containing RNA-sequencing data recalculated into transcripts per million, was obtained from the UCSC Xena platform. Three thousand nine hundred and seventy six differentially expressed genes were obtained with analysis of variance. Using these data a co-expression network was constructed using weighted gene co-expression network analysis, from which we obtained eight modules. Results: The blue module included 497 genes and demonstrated significant negative correlation with overall survival. Furthermore, pathway analyses demonstrated the involvement of many of these genes in the tight junction pathway, which plays a critical role in PAAD. In addition, we identified six genes in common (i.e., ANXA2 [annexin A2], EPHA2 [erythropoietin-producing hepatocellular class A2], ITGB4 [integrin beta 4], KRT19 [keratin type I cytoskeletal 19], LGALS3 [galectin-3], and S100A14 [S100 calcium binding protein A14]) between the protein-protein interaction and gene co-expression networks that may have critical functions in PAAD. These hub genes were not only highly expressed at the RNA level but also exhibited high expression in the immunohistological data in the Human Protein Atlas Database. Conclusion: Thus, this research clarified the framework of co-expressed gene modules in PAAD and highlighted potential prognostic biomarkers for the clinical diagnosis of PAAD.
Keywords: PAAD; WGCNA; hub genes; prognosis; tight junctions.